Download the influence of a professional organization on physician behavior

THE INFLUENCE OF A PROFESSIONAL
ORGANIZATION ON PHYSICIAN BEHAVIOR
David Orentlicher*
INTRODUCTION
As many commentators have observed, our society often operates
on the basis of the technological imperative: advances in medicine
and in other fields are used because they are available, for their own
sake, rather than for the overall good of the public.1 For example,
cardiopulmonary resuscitation ("CPR") is sometimes administered to
patients for whom there is virtually no possibility of restoring cardiac function, simply on the presumption that CPR should be applied to any patient whose heart has ceased beating.2
Medical innovations may also be misused on account of invidious
biases. Tests for genetic abnormalities have been used in the past as
a basis to unfairly discriminate against individuals who carry the
abnormalities. 3 Other inappropriate biases may also drive the use of
innovations. For example, excessive resources have been devoted to
certain technologies because of a societal bias in favor of treatment
4
rather than prevention.
To ensure that medical innovations are used in an ethically responsible fashion, a variety of approaches have been used. Professional societies establish guidelines,5 insurers employ utilization
* Ethics and Health Policy Counsel, American Medical Association; Lecturer in Law,
University of Chicago Law School; Adjunct Assistant Professor of Medicine, Northwestern
University Medical School. A.B., Brandeis University, 1977; M.D., Harvard University, 1981;
J.D., Harvard University, 1986. The views expressed herein do not necessarily represent
those of the American Medical Association.
1 See, e.g., Eric J. Cassell, The Sorcerer'sBroom: Medicine's Rampant Technology, HASTINGS
CENTER REP., Nov.-Dec. 1993, at 32; Nancy K. Rhoden, LitigatingLife and Death, 102 HARv. L.
REv. 375, 420-27 (1988).
2 See Rhoden, supra note 1, at 423; see also Paul C. Sorum, Limiting Cardiopulmonary
Resuscitation, 57 ALB. L. REV. 617, 618 (1994) (arguing that "limiting [CPR's] use must depend
on patient choice").
3 Council on Ethical and Judicial Affairs, American Medical Association, Use of Genetic
Testing by Employers, 266 JAMA 1827, 1827 (1991).
4 TOM L. BEAucHAMP & JAMES F. CHILDRESS, PRINCIPLES OF BIOMEDICAL ETHICS 288 (3d ed.
1989).
5 See, e.g., Subcommittee on Pacemaker Implantation, Joint American College of
Cardiology/American Heart Association Task Force on Assessment of Cardiovascular
Procedures, Guidelinesfor Permanent CardiacPacemakerImplantation,May 1984, 4 J. AM. C.
CARDIOLOGY 434 (1984).
Albany Law Review
584
[Vol. 57
review, 6 and governments adopt legal rules. 7 My purpose is not to
determine which of these approaches is most valuable. There is undoubtedly a role for all of them, with their importance depending
upon the specific innovation and the specific use of the innovation in
question. Indeed, a combination of approaches will often make the
most sense.
I would like to make three points in this Article. First, professional societies can play, and indeed have played, an important role
in establishing guidelines for ethical physician practices. Second,
merely setting standards alone is insufficient to shape physician behavior; the standards generally need to be supplemented by other
measures or incentives. Third, professional societies are much more
successful at setting standards than they are at ensuring physician
acceptance and implementation of the standards. The kinds of incentives or mandates that are necessary for the adoption of ethical
standards generally have come from outside the profession.
In developing these points, I will refer to empirical studies of efforts to regulate physician behavior, as well as my own experiences
in trying to shape physician behavior through the issuance of ethical
guidelines by the American Medical Association ("AMA"). But
before getting into the specific examples, I will discuss why it is important for the medical profession to have some responsibility for establishing guidelines on ethical matters.
I.
PROFESSIONAL RESPONSIBILITY
There are several reasons why the medical profession ought to
have an important role in the setting of guidelines on ethical
questions.
A.
ProfessionalAutonomy
In the past two or three decades, patient autonomy has become
the dominant ethical principle in medicine.'
Patient selfdetermination has replaced beneficence by the physician as the fun6 See Theodore R. Marmor & Michael S. Barr, Making Sense of the National Health
Insurance Reform Debate, 10 YALE L. & POL'Y REV. 228, 232 (1992) ("Private insurance firms
spend large and increasing sums on utilization reviews, marketing, and billing.").
7 For example, Medicare denies reimbursement for cellular therapy because it "is without
scientific or statistical evidence to document its therapeutic efficacy and, in fact, is considered
a potentially dangerous practice." 5 Medicare & Medicaid Guide (CCH) 27,201, at 29,215
(1993).
8 See BEAUCHAMP & CHILDRESS, supra note 4, at 67-113.
1994]
Influence of a Professional Organization
damental value in medical decision making.9 Indeed, as David
Blake has observed, when courts are faced with a patient's request
to discontinue life-sustaining treatment, they almost uniformly give
lip service to the state's interests in continuing treatment, and ultimately defer to the patient's preferences. 10 In the liberal state, in
which there is no universal consensus on what is good, individuals
must be given considerable freedom from externally imposed moral
values. It is incumbent upon the state to maintain moral neutrality
and permit individuals to define their own sense of morality, as long
as there is no infringement on the freedom of fellow citizens.11
Physicians also have a strong need for personal autonomy. Society
respects individual dignity by permitting people to have control over
essential aspects of their lives, and for most people, professional expression is a critical element of personhood. As Robert Gordon observed, control over the working environment is a basic
"precondition to the realization of [a] free, authentic personality."12
To be sure, within the liberal state there may be subcommunities of
persons who share a common moral view,13 and physicians may each
therefore agree to be bound by a professional code of ethics. Nevertheless, principles of personal self-expression indicate that it is still
for the profession, rather than the state, to develop a professional
code.
Patients also benefit from physician autonomy. Medicine will not
attract talented individuals if physicians are not given the opportunity for self-regulation. Currently, in explaining an increasing disenchantment with their professional lives, physicians identify loss of
autonomy as one of the most important problems.14 Doctors express
grave concern about mounting paperwork, heavyhanded utilization
review, and greater government regulations-the so-called "hassle
factor."15 Such a reaction is predictable. As studies have demonstrated, when individuals exercise more control over decision mak9 See id. at 210 ("So influential is th[e] autonomy model at the present time that it has
become difficult to find clear commitments to the traditional beneficence model in
contemporary biomedical ethics.").
10 David C. Blake, State Interests in Terminating Medical Treatment, HASTINGS CENTER
REP., May-Jun. 1989, at 5.
11 See H. TRISTRAM ENGELHARDT, JR., THE FOUNDATIONS OF BIOETHICS 17-56 (1986).
12 Robert W. Gordon, The Independence of Lawyers, 68 B.U. L. REV. 1, 9 (1988).
13 ENGELHARDT,
supra note 11, at 49-56.
Arnold M. Epstein, Changes in the Delivery of Care Under Comprehensive Health Care
Reform, 329 NEW ENG. J. MED. 1672, 1673 (1993).
14
15
Id.
Albany Law Review
[Vol. 57
ing in their workplace, they experience16 greater satisfaction and are
more productive in their employment.
B.
Moral Responsibility
Professional autonomy enhances the physician's sense of moral responsibility. "By collectively engaging in the process of enacting and
enforcing" guidelines on ethical questions, physicians, like other professionals, "develop and reinforce [a] disposition for moral decisionmaking. " 1 7 Society has a profound interest in "cultivating a moral
personality" in physicians, as with all of its citizens.' 8
If physicians are responsible for establishing their own ethical
code, they are much more likely to view ethics as an integral part of
the practice of medicine. For decades, the public has called on the
profession to integrate ethics more fully into medical training and
practice.' 9 While ethics instruction is now offered in all medical
schools, it generally is treated as a supplemental rather than core
element of the curriculum;2 ° there is still a need for greater emphasis on the incorporation of ethics into medicine. If we say that ethical standards are not really the responsibility of the profession, but
of other groups, then we are essentially divorcing ethics from the
practice of medicine.
C.
Independence from Political Processes
While professional self-regulation has inherent conflicts of interest and may be compromised by political pressures, too great a reliance on government regulation risks even greater politicization of
the standard-setting process. Examples from the past twelve years
are illustrative. Political considerations prevented the Reagan and
Bush Administrations from even convening a federal ethics commis16 R.J. BULLOCK, IMPROVING JOB SATISFACTION 4-5 (1984); Katherine I. Miller & Peter R.
Monge, Participation,Satisfaction, and Productivity:A Meta-Analytic Review, 29 AcAD. MGMT.
J. 727 (1986).
17 See David B. Wilkins, Who Should Regulate Lawyers?, 105 HARv. L. REV. 799, 863 (1992)
(footnote omitted) (reaching this conclusion in the context of the legal profession).
18 See id.
19 In fact, in 1985 it was noted that "[iun scarcely more than a decade, the teaching of
medical ethics has become a regular feature of medical school curricula in the United States."
Edmund D. Pellegrino et al., Relevance and Utility of Courses in Medical Ethics: A Survey of
Physicians'Perceptions, 253 JAMA 49, 49 (1985).
20 See id. Additionally, "the majority of residency training programs in internal medicine
are just beginning to try to accommodate the need for training in the ethical aspects of patient
care." Daniel P. Sulmasy et al., Medical House Officers'Knowledge, Attitudes, and Confidence
Regarding Medical Ethics, 150 ARCHIVES INTERNAL MED. 2509, 2509 (1990).
1994]
Influence of a Professional Organization
sion, despite congressional authorization, 2 1 and delayed for more
than four years the implementation of recommendations by a federal
advisory panel on fetal tissue transplantation.2 2 Even today it is
highly unlikely that the U.S. Congress would reach the same position that the AMA's Council on Ethical and Judicial Affairs did when
it concluded that parental involvement should not be mandatory
when minors have abortions.23
In some cases, independent government commissions can be used
to insulate standard setters from political influences. Difficult political questions, like the closing of military bases, are often resolved
by the establishment of "blue-ribbon" panels that make recommendations to Congress after careful study.2 4 However, the recent
politicization of Supreme Court appointments 25 reflects how difficult
it is to maintain the independence and integrity of government bodies that have a continuing responsibility.
There is a real danger of overreaching when the government establishes guidelines on ethical issues; the government may be easily
tempted to use its ethics pronouncements to serve other policy goals.
Indeed, last year, when it issued its proposed regulations for executions, the federal government planned to have physicians participate
in capital punishment. 26 As a result of objection from the medical
profession, however, the final rules excluded any requirement that
physicians be involved.2 7
As the government is becoming both increasingly involved in the
provision of health care and increasingly concerned about rising
health care costs, 28 intense pressure will develop to sacrifice ethical
values for economic reasons. There is already evidence that the federal government is insufficiently concerned about financial incentives by health care plans that discourage physicians from using
diagnostic and therapeutic procedures. For example, health mainte21 OFFICE OF TECHNOLOGY ASSESSMENT, U.S. CONGRESS, BIOMEDICAL ETHICS IN U.S. PUBLIC
POLICY-BACKGROUND PAPER 11 (1993).
22 Id. at 9.
23 Council on Ethical and Judicial Affairs, American Medical Association, Mandatory
Parental Consent to Abortion, 269 JAMA 82 (1993).
24 See Eric Schmitt, A Mission Accomplished, N.Y. TIMES, June 29, 1993, at A10.
25 See, e.g., Survey, The Supreme Court Appointment Process, 57 ALB. L. REV. (forthcoming
June 1994).
26 Implementation of Death Sentences in Federal Cases, 57 Fed. Reg. 56,536, 56,537 (1992)
(to be codified at 28 C.F.R. pt. 26) (proposed Nov. 30, 1992).
27 Implementatio n of Death Sentences in Federal Cases, 28 C.F.R. § 26.4 (1993); see
Andrew A. Skolnick, Health Professionals Oppose Rules Mandating Participation in
Executions, 269 JAMA 721 (1993).
28 See, e.g., Adam Clymer, President Seeks to 'Write a New Chapter in the American Story,'
N.Y. TIMES, Sept. 23, 1993, at Al.
Albany Law Review
[Vol. 57
nance organizations and other managed care plans typically use fee
withholds 2 9 and bonuses to reward physicians more when their patients consume fewer health care resources.3 0 Accordingly, physicians recognize that every time they order a test or a procedure,
their income will be reduced. The Health Care Financing Administration has proposed rules to limit the amount of a physician's income that can be placed at risk, but the rules would still allow fee
withholds and bonuses to account for up to 30% of a physician's income. 3 1 When 30% of a physician's income is at risk, health care
costs may be contained, but the possibility that necessary treatment
will be withheld becomes very problematic. In short, governmental
bodies that establish and enforce ethics guidelines may have a tendency to sacrifice important ethical principles to achieve political
goals.
Professional standard-setting bodies can also become politicized.
In the case of the AMA's Council on Ethical and Judicial Affairs,
however, several steps have been taken to minimize that risk. First,
except for the medical student and the resident, members of the
Council serve a single seven-year term (the student and resident
members generally serve a single term of two or three years, respectively).32 Accordingly, the decisions of Council members are not influenced by their concerns about reappointment. Second, Council
members are nominated by the President of the AMA and approved
by the House of Delegates. 3 Because the members do not run for
office, they are not under the pressure of an election campaign to
pander to sentiments that are popular but not ethically meritorious.
Third, while on the Council, members may not hold any other positions in the AMA. 3 4 If a person has a position in the House of Delegates or on the Board of Trustees when appointed to the Council, the
position must be relinquished. Consequently, individuals with
strong political ambitions do not seek positions on the Council; a
29 A fee withhold refers to the practice of health care plans withholding a percentage of the
physician's compensation at each pay period and giving all or part of the withheld payments to
the physician at a later date, depending on the amount of services used by patients in the plan.
Medicare and Medicaid Programs; Requirements for Physician Incentive Plans in Prepaid
Health Care Organizations, 57 Fed. Reg. 59,024, 59,026 (1992) (to be codified at 42 C.F.R. pts.
417, 434, 1003) (proposed Dec. 14, 1992).
30 See id. at 59,025 to 26.
31 Id. at 59,032.
32 COUNCIL ON ETHICAL AND JUDICIAL AFFAIRS, AMERICAN MEDICAL ASSOCIATION, 1992 CODE
OF MEDICAL ETHICS-ANNOTATED CURRENT OPINIONS 86-88 (1992) [hereinafter CODE OF
ETHics].
33 Id. at 87.
34 Id.
1994]
Influence of a Professional Organization
seven-year hiatus from the political process generally would preclude future opportunities for political leadership in the AMA. Because Council members have few political aspirations, they are not
as likely to temper their opinions with political concerns. Fourth,
the Council has the authority to issue positions on ethical issues
either with or without approval of the House of Delegates.8 5 While
House approval is often sought to avoid a sense that the Council is
ramming positions through in an authoritarian fashion, the Council
routinely issues guidelines without House approval and even over
House objection, if necessary. For example, after the House twice
rejected a Council report that called for specific informed consent as
a prerequisite to HIV testing, the Council, without seeking House
approval, issued an opinion that imposed the requirement of specific
consent.36
D. Acceptability
Ethical guidelines will be more readily accepted by physicians if
they are developed internally rather than if they are imposed by external bodies, particularly external bodies with a poor record of selfregulation. For example, it is easy for physicians to dismiss calls by
Congress for stricter rules on conflicts of interest in medicine when
members of Congress routinely accept contributions, invitations to
visit resorts, and other gifts from businesses that are affected by potential legislation.
The public also appears to hold more trust in the medical profession than in other bodies when it comes to setting ethical standards.
In polls taken over the past fifteen years, 50-60% of those surveyed
gave a "high" or "very high" rating to physicians on both honesty and
ethical standards.3 7 In contrast, only 10-25% of the respondents
rated United States Senators as having "high" or "very high" ethical
The comparable figures for United States Congressstandards.3
men were 10-20%. 9 Similarly, on the question of who the public
35 Council on Ethical and Judicial Affairs & Council on Constitution and Bylaws, American
Medical Association, Clarification of House Procedures with Respect to the Opinions and
Reports of the Council on Ethical and JudicialAffairs, in PROCEEDINGS OF THE HOUSE OF
DELEGATES, 45TH INTERIM MEETING
330-33 (1991).
36 CODE OF ETHICS, supra note 32, at 26. The House preferred the view that consent to HIV
testing be implied in a patient's general consent to necessary blood tests. AMERICAN MEDICAL
ASSOCIATION, POLICY COMPENDIUM § 20.945 (1993).
37 Leslie McAneny, Honesty and Ethics Poll: Pharmacists Retain Wide Lead as Most
Honorable Profession, L.A. Times Syndicate, July 29, 1993, available in Gallup Poll News
Service.
38 Id.
39 Id.
Albany Law Review
[Vol. 57
trusts for proposing fair and workable health policies, 71% of those
surveyed trust organized medicine "some" or "a great deal" while
40
only 40% trust the Federal Government "some" or "a great deal."
E.
Reliability
In discussions about self-regulation in medicine, it is commonly
argued that lay people do not have sufficient expertise to set standards for medical practice while physicians are much better
equipped to establish practice guidelines. 4 1 For example, the medical profession establishes standards for determining whether a physician has practiced competently.4 2 While it is true that ethical
guidelines may not require the same degree of professional expertise
as technical guidelines, physicians do have unique perspectives on
the patient-physician relationship that are critically relevant in determining what guidelines are appropriate. Similarly, physicians
are often in a better position than lay persons to make ethical judgments because of the importance of medical facts to ethical conclusions. For example, when deciding whether a child should serve as a
donor in a kidney transplantation, it is necessary to understand the
risks to the donor and the benefits to the recipient. In addition, because the ethics community has yet to develop its own body of standards, it is not clear that ethicists are in any better position to
develop guidelines for the profession.
F.
Due Process
According to principles of due process, the medical profession
ought to have the- opportunity to resolve ethical problems itself
before other groups jump in with externally developed mandates.
Due process recognizes that in ensuring a just outcome the mechanism for achieving a result is as important as the result itself.4 3 Indeed, the tradition of oral argument in appellate litigation is
important not so much because it affects the court's decision but because oral argument assures the litigants that they themselves have
been heard. 44 Consider also that courts generally do not review ad40 LYNN K. HARVEY, AMA SURVEY OF PUBLIC OPINION ON HEALTH CARE ISSUES 18 (1991).
41 See W. PAGE KEETON ET AL., PROSSER AND KEETON ON THE LAW OF TORTS § 32, at 189 (5th
ed. 1984).
42 Id. at 186-89.
43 See JOHN E. NowAK & RONALD D. ROTUNDA, CONSTITUTIONAL LAw § 13.1, at 487 (4th ed.
1991) ("The due process clauses also have a procedural aspect in that they guarantee that each
person shall be accorded a certain 'process' if they are deprived of life, liberty, or property.").
44 See, e.g., ROBERT L. STERN, APPELLATE PRACTICE IN THE UNITED STATES § 13.1, at 365-70
(2d ed. 1989).
1994]
Influence of a Professional Organization
ministrative agency decisions until challenges have been taken
through administrative appeal (i.e., administrative remedies have
been exhausted). 4 5 Requiring the exhaustion of administrative remedies provides administrative agencies the opportunity to correct
their mistakes before courts step in. 46 Just as administrative agencies are given a chance to internally correct errors, so should the
medical profession be permitted to correct ethical problems through
self-regulation before others step in with their own approaches.
II.
CASE STUDIES
Giving theoretical arguments to justify professional selfregulation of ethics is only part of the story. We still need to know
whether professional self-regulation actually works. Can the profession engage in meaningful self-regulation, or is this just a matter of
letting foxes guard the chicken coop? The medical profession's experience with standard-setting suggests the following conclusion: professional regulation can have a substantial impact on physician
behavior, but professional guidelines alone are generally insufficient
to change physician behavior. The guidelines must be combined
with other measures to ensure compliance.
I will discuss two sources of data that can inform us on the effectiveness of professional standards on ethical issues: the impact of
the AMA's guidelines on ethical questions and the impact of professional guidelines for therapeutic procedures such as guidelines that
indicate when cesarean sections and coronary artery bypass surgery
should be performed. The experience with the AMA's ethical guidelines is directly relevant. However, an assessment of their impact
must rely in part on anecdotal data; well-controlled studies of their
impact do not exist. Professional guidelines on therapeutic procedures are less relevant because they involve decisions that are both
technical and ethical in nature. On the other hand, there have been
many studies examining the impact of these guidelines, and we can
safely assume that some of the lessons learned apply to the implementation of ethical guidelines.
45 See, e.g., Myers v. Bethlehem Shipbuilding Corp., 303 U.S. 41, 50-51 (1938) (noting "the
long settled rule ofjudicial administration that no one is entitled to judicial relief... until the
prescribed administrative remedy has been exhausted").
46 2 KENNETH C. DAVIS & RICHARD J. PIERCE, JR., ADMINISTRATIVE LAW TREATISE § 15.2, at
309 (3d ed. 1994).
[Vol. 57
Albany Law Review
A.
The AMA's Ethical Guidelines
The AMA's Council on Ethical and Judicial Affairs issues a Code of
Ethics 47 for physicians that is analogous to the ABA's Model Code of
Professional Responsibility.48 Physicians who violate the AMA's
code are subject to discipline by the AMA, and by their county and
state medical societies. 49 A number of specialty societies, including
the American Academy of Family Physicians and the American Psychiatric Association, have adopted the AMA's code and hold their
members accountable for violations.50 In some states, the medical
licensing statute expressly considers violations of the AMA's code as
grounds for discipline.51 Apparently, state licensing boards generally view the AMA's code as probative, though not dispositive, evidence of the expected standard of conduct when deciding whether a
physician has committed professional misconduct.52
From my five years as Secretary to the Council on Ethical and Judicial Affairs, I will discuss an example of successful self-regulation
and an example of unsuccessful self-regulation and suggest why the
two efforts had different results.
1. Gifts to Physicians from Industry
During the 1980s, there was increasing concern in the medical
profession about gifts to physicians from pharmaceutical and other
companies.53 Commentators were troubled both by the magnitude
and kinds of industry gift-giving.54 Data on magnitude was devel47 See CODE OF ETHICS, supra hote 32. Revised versions of the code are issued every two to
three years. The 1994 edition is scheduled for publication in May 1994.
48 MODEL CODE OF PROFESSIONAL RESPONSIBILITY
49 See CODE
OF ETHICS,
(1980).
supra note 32, at 1, 78-79, 94-98 (§§ 1.01, 1.02, and 9.04 of the Code
of Ethics and describing disciplinary procedures).
50 See, e.g., AMERICAN ACADEMY OF FAMILY PHYSICIANS,
PSYCHIATRIC ASSOCIATION, THE PRINCIPLES
ESPECIALLY APPLICABLE To PSYCHIATRY (1989).
OF MEDICAL
BYLAws 6-7 (1993); AMERICAN
ETHICS:
WITH
ANNOTATIONS
51 See, e.g., OHIO REV. CODE ANN. § 4731.22(B)(18) (Anderson Supp. 1992).
52 Telephone interview with Dorothy Harwood, Assistant Vice President for Administrative
and Legislative Affairs, Federation of State Medical Boards of the United States (Feb. 3,
1994).
53 See, e.g., Mary-Margaret Chren et al., Doctors, Drug Companies, and Gifts, 262 JAMA
3448 (1989); Stephen E. Goldfinger, A Matter of Influence, 316 NEW ENG. J. MED. 1408 (1987);
Michael D. Rawlins, Doctors and the Drug Makers, LANCET, Aug. 4, 1984, at 276.
64 See, e.g., Council on Ethical and Judicial Affairs, American Medical Association, Gifts to
Physiciansfrom Industry, 265 JAMA 501 (1991) (noting that "there has been growing concern
about certain gifts from industry to physicians"); John Graves, Frequent-flyer Programs for
Drug Prescribing, 317 NEW ENG. J. MED. 252 (1987) (letter to the editor); Teri Randall,
Kennedy HearingsSay No More Free Lunch--or Much Else-FromDrug Firms, 265 JAMA 440
(1991) (noting Senator Kennedy's remark: "'Doctors who accept lavish industry gifts are
jeopardizing their objectivity and compromising the trust of their patients' ").
1994]
Influence of a Professional Organization
oped by the Senate Labor and Human Resources Committee, which
tracked expenditures by eighteen large pharmaceutical companies
on gifts to physicians between 1975 and 1988. 5s Over that period,
after taking inflation into account, gift expenditures nearly quintupled. 56 There also appeared to be a greater tendency for companies to give gifts particularly likely to influence the treatment
decisions of physicians. Gift-giving extended well beyond pens,
mugs, and grants for educational programs to all-expense paid
57
weekend trips at lavish resorts for physicians and their spouses,
frequent prescriber programs offering free airline tickets for every
fifty prescriptions, and "studies" which paid physicians hundreds
of dollars if they prescribed expensive antibiotics and collected data
that was essentially demographic in nature.5 9
By 1990, guidelines on gift-giving had been issued by a number of
professional societies, including England's Royal College of Physicians, 60 the American College of Physicians,6 1 and the American College of Cardiology.6 2 However, there was little evidence of change in
industry gift-giving practices. While praiseworthy, the guidelines
lacked specificity. For example, physicians were admonished to decline gifts that they were not willing to have "generally known" to
others.6 3 This vagueness made it difficult to charge anyone with violations of these guidelines.
Following nearly a year of deliberations, the AMA issued its own
guidelines on gift-giving in December 1990.64 The guidelines explicitly prohibit cash payments, subsidies for the travel expenses of physicians attending conferences, gifts tied to prescribing practices, and
any gift not related to patient care. 5 In addition, the guidelines
limit the magnitude of individual gifts and require that grants to
5 Randall, supra note 54, at 440, 442.
56 Id. at 442.
57 John C. Nelson, A Snorkel, a 5-Iron, and a Pen, 264 JAMA 742 (1990).
58 See Graves, supra note 54.
59 Randall, supra note 54, at 440.
60 Royal College of Physicians,
The Relationship between Physicians and the
PharmaceuticalIndustry, 20 J. ROYAL C. PHYsIcIANs LONDON 235 (1986).
61 American College of Physicians, Physicians and the Pharmaceutical Industry, 112
ANNALs INTERNAL MED. 624 (1990).
62 C. Richard Conti et al., Task Force V: The Relation of CardiovascularSpecialists to
Industry, Institutions and Organizations, 16 J. AM. C. CARDIOLOGY 30 (1990).
63 See American College of Physicians, supra note 61, at 624; C. Richard Conti et al., supra
note 62, at 32; Royal College of Physicians, supra note 60, at 238.
64 Council on Ethical and Judicial Affairs, American Medical Association, supra note 54.
65 See CODE OF ETmics, supra note 32, at 68-69; Council on Ethical and Judicial Affairs,
American Medical Association, supra note 54.
Albany Law Review
[Vol. 57
defray registration fees for educational conferences
be given directly
66
physicians.
to
not
and
to conference sponsors
Ordinarily, it is difficult to measure the impact of ethical guidelines. It is not always certain whether ethical guidelines result in
behavioral changes. Even when changes are detected, it is often not
clear whether the changes reflect the ethical guideline or other contemporaneous influences. For example, if there is an increase in
services provided to the poor after the issuance of a guideline calling
on physicians to care for the indigent, the increase may be the result
of the ethical guideline or perhaps, the result of a coincidental rise in
Medicaid reimbursement rates.
With the AMA gift-giving guidelines, however, the impact was immediate and substantial. Companies canceled educational and promotional conferences that were not strong enough to attract
physicians willing to pay their own travel expenses, and promotional
dinners where physicians received a free meal and a $100 payment
were also abandoned. 67 At the Council on Ethical and Judicial Affairs, we received calls from travel agencies complaining about the
impact of the gift-giving guidelines on their businesses, and physicians reported that lavish evening receptions were disappearing at
major medical meetings. In this case, the ethical guidelines changed
physician behavior dramatically and meaningfully.
Why were these guidelines so successful? First, the pharmaceutical industry incorporated the guidelines into its ethics code for marketing practices.6 8 As a result, the success of the guidelines was not
solely dependant on the willingness of physicians to adhere to their
ethical responsibilities. After implementation, drug companies generally stopped offering inappropriate gifts; thus physicians were not
in a position to accept them. In fact, the industry probably did not
fight the guidelines too vigorously because in some ways, companies
welcomed the restrictions. To a certain extent, gifts are given because of physician demand, 69 and once one company accedes to such
demands, other companies must follow in order to remain competitive. Similarly, one company may initiate gift-giving as a marketing
strategy and other companies, for competitive reasons, feel compelled to match the strategy. A prohibition on gift-giving levels the
66 CODE OF ETHICS, supra note 32, at 68-69; Council on Ethical and Judicial Affairs,
American Medical Association, Guidelines on Gifts to Physiciansfrom Industry: An Update, 47
FOOD DRUG L. J. 445, 452 (1992).
67 Teri Randall, AMA, Pharmaceutical Association Form 'Solid Front' on Gift-Giving
Guidelines, 265 JAMA 2304 (1991).
68 Id.
69 Royal College of Physicians, supra note 60, at 237.
1994]
Influence of a Professional Organization
595
playing field for members of the industry, having the same effect as
an agreement by the companies that they would not try to compete
with each other through gift-giving.7 0
Second, while there are more than one hundred drug companies, a
small number of large companies dominate the market.7 1 Consequently, in order to ensure that the guidelines achieved their purpose, it was necessary to achieve compliance from only a few major
companies. Moreover, with the focus on just the dominant players in
the industry, policing the guidelines became much easier as well.
Third, detection of violations is relatively easy. Gift-giving occurs
openly, and companies usually offer the same gift to hundreds, if not
thousands of physicians. Physicians who support the guidelines as
well as competitors of the gift-giving company are likely to be aware
of violations and report them to the AMA. Indeed, in the months
following the implementation of the guidelines, our attention was
drawn to a number of apparent violations.7 2
Fourth, concern about government regulation gave both physicians and industry a strong incentive to follow the AMA's guidelines.
Immediately following the issuance of the guidelines, Senator
Edward Kennedy convened hearings on the pharmaceutical industry's gift-giving practices. 73 After the hearings, he indicated that he
would refrain from taking any legislative or regulatory efforts if the
AMA's guidelines eliminated abusive gift-giving practices. 74 Thus,
even though there was not a strong enforcement mechanism in
place, 7 5 there was a strong threat of enforcement by means of legislative action looming over both the pharmaceutical industry and the
medical profession.
Fifth, the, guidelines draw a number of "bright line" rules, establishing clear distinctions between permissible and impermissible
conduct.76 The pharmaceutical industry had previously adopted the
American College of Physician's guidelines on gift-giving practices,
70 Such an agreement would, of course, be unlawful under antitrust law. See 15 U.S.C. § 1
(1988). Agreeing to the AMA's code could also constitute a violation of antitrust law, but the
risk of prosecution is very low.
71 See ARSEN J. DARNAY, MARKET SHARE REPORTER: AN ANNUAL COMPILATION OF REPORTED
MARKET SHARE DATA ON COMPANIES, PRODUCTS, AND SERVICES
72 Randall, supra note 67.
191-92 (1991).
73 Randall, supra note 54.
74 Randall, supra note 67, at 2305.
75 Randall, supra note 54, at 442.
76 CODE OF ETHICS, supra note 32, at 68-69 (setting forth the provision relating to "Gifts to
Physicians from Industry").
Albany Law Review
[Vol. 57
but, because those guidelines essentially enunciated general principles, industry had a good deal of freedom in interpreting them.7 7
In short, the AMA gift-giving guidelines probably succeeded because the rules were clear, 78 because they actually served the interests of one of the parties effected, because there was a credible
threat of enforcement in the form of greater government oversight,
and because violations could be detected with relative ease.
2.
Treatment of HIV-Infected Patients
There has apparently been less success with the AMA's ethical
guideline on the duty of physicians to treat patients with HIV infection. In December 1987, the Council on Ethical and Judicial Affairs
issued a guideline stating that physicians may not refuse to treat
patients on account of their HIV infection.79 Since then, however,
studies suggest that a substantial number of physicians have not followed the guideline.80 In an August 1990 random national sample of
primary care physicians, 50% of the physicians surveyed stated that,
if given a choice, they would not work with AIDS patients, and 48%
stated that they preferred to refer patients with HIV infection to
other physicians.8 ' Similarly, in a survey of one thousand surgeons,
more than 90% expressed support for a policy of refusing to operate
on patients with HIV infection.8 2 Since these surveys report attitudes rather than actual practices, it is possible that the surveyed
physicians overcame their unwillingness to treat patients with HIV
infection and hewed to their ethical responsibilities. Indeed, a 1986
survey of orthopedic surgeons suggested that while more than twothirds of orthopedists believed that a surgeon could ethically refuse
to operate on a patient with HIV infection, 90% of the orthopedists
who had an opportunity to operate on infected patients had done so
on at least one patient with HIV infection. 3
77 See Randall, supra note 54, at 442.
78 Clear guidelines were also a key factor in the success of a federal government regulation
limiting the use of antipsychotic drugs in nursing homes. Robert L. Kane & Judith Garrard,
Changing Physician Prescribing Practices:Regulation vs. Education, 271 JAMA 393, 393
(1994).
79 CODE OF ETHICS, supra note 32, at 85; Council on Ethical and Judicial Affairs, American
Medical Association, Ethical Issues Involved in the Growing AIDS Crisis, 259 JAMA 1360
(1988).
80 See, e.g., ACLU AIDS PROJECT, EPIDEMIC OF FEAR: A SURVEY OF AIDS DISCRIMINATION IN
THE 1980s AND POLICY RECOMMENDATIONS FOR THE 1990s 78-80 (1990).
81 Barbara Gerbert et al., Primary Care Physicians and AIDS: Attitudinal and Structural
Barriersto Care, 266 JAMA 2837, 2839 (1991).
82 ACLU AIDS PROJECT, supra note 80, at 80.
83 Paul M. Arnow et al., Orthopedic Surgeons' Attitudes and Practices Concerning
Treatment of Patients with HIV Infection, 104 PUB. HEALTH REP. 121, 124, 127 (1989). Of
1994]
Influence of a Professional Organization
Several other studies, however, indicate that actual practices deviate from the ethical duty to treat. In a survey of Los Angeles County
primary care physicians in late 1990, researchers found that 48% of
the physicians surveyed had either refused or would refuse to accept
HIV infected patients into their practice. 4 Similarly, in a June 1990
survey of North Carolina physicians, 40% reported that they either
refused to treat HIV-infected patients or referred the patients elsewhere.85 In a 1989 survey of resident physicians, 39% of those surveyed in the United States reported that at least one of their HIVinfected patients had been refused care by a surgeon. 6 Finally, a
1989 national survey of 560 randomly selected hospitals found that
20% of these hospitals had experienced at least one case of a staff
member refusing to treat a patient with HIV infection; similarly,
25% of these hospitals immediately transferred HIV-infected patients to other hospitals.8 7
Why has there been less success with the guideline on the duty to
treat patients with HIV infection than with the guideline on gifts
from industry? A number of possible explanations come to mind.
First, there are strong personal incentives to ignore the obligation to
provide treatment. Physicians, particularly surgeons, are concerned
that they will become infected from HIV patients while treating
them. 8 While the perceived risk may be greater than the actual
risk, it is perceptions that drive behavior. Physicians may also be
discouraged from treating HIV-infected patients because of the psychological burdens of providing care. That is, because of the difficult
clinical course, caring for HIV-infected patients is often timeconsuming and emotionally draining.
Second, it is easy to camouflage violations of the obligation to
treat. Physicians who do not want to treat a patient with HIV infection can simply tell the patient that they are not taking any new
course, it is possible that many of the 90% refused to treat the majority of HIV-infected
patients who sought care from them.
84 Charles E. Lewis & Kathleen Montgomery, Primary Care Physicians'Refusal to Carefor
PatientsInfected with the Human Immunodeficiency Virus, 156 W. J. MED. 36, 37 (1992).
85 Morris Weinberger et al., Physicians' Attitudes and PracticesRegarding Treatment of
HIV-Infected Patients, 85 S. MED. J. 683, 685 (1992).
86 Martin F. Shapiro et al., Residents' Experiences in, and Attitudes Toward, the Care of
Persons With AIDS in Canada, France,and the United States, 268 JAMA 510, 512 (1992).
87 Philip J. Hilts, Many HospitalsFound to Ignore Rights of Patients in AIDS Testing, N.Y.
TIMES, Feb. 17, 1990, at Al, A12.
8 Weinberger et al., supra note 85, at 684.
Albany Law Review
[Vol. 57
patients, or that they accept patients only through a referral.8 9
Moreover, even when violations are detected, there may not be a
credible threat of enforcement. Currently, the Americans with Disabilities Act ("ADA) 9 ° prohibits physicians from denying care to patients on account of their HIV infection. 9 1 However, until the ADA
went into effect in July of 1992, state anti-discrimination laws provided weak protection for patients with HIV infection.9 2
From these two examples of ethics guidelines, we can take away
two important points. First, the medical profession is perfectly capable of devising meaningful and responsible guidelines on ethical
matters, even when guidelines require conduct that might not be in
the physician's own personal interest. Second, the profession is less
successful when it comes to ensuring that guidelines are followed.
Consequently, guidelines will probably not be adopted in practice
unless there is some credible method of enforcement from outside
the profession. As discussed in the next section, these two lessons
can also be derived from the medical profession's experience with
practice guidelines.
B.
Practice Guidelines
To ensure that physician practices are consistent with quality
medical care, professional societies have developed standards of
practice for a wide range of clinical situations. The American Academy of Pediatrics has published schedules for childhood vaccinations,9 3 the American College of Cardiology has issued guidelines for
exercise testing,9 4 coronary artery bypass surgery,9 5 and pacemaker
implantation,9 6 and the American Society of Anesthesiologists has
89 See Mark H. Jackson & Nan D. Hunter, "The Very Fabric of Health Care": The Duty of
Health Care Providersto Treat People Infected with HIV, in AIDS AGENDA: EMERGING ISSUES
IN CIL RIGHTS 123, 124 (Nan D. Hunter & William B. Rubenstein eds., 1992).
90 Americans with Disabilities Act of 1990, 42 U.S.C. §§ 12,101-12,213 (Supp. III 1991).
91 Jackson & Hunter, supra note 89, at 130.
92 ACLU AIDS PROJECT, supra note 80, at 64-66.
93 AMERICAN ACADEMY OF PEDIATRICS, REPORT OF THE COMMITTEE ON INFECTIOUS DISEASES
5-60 (21st ed. 1988).
94 Subcommittee on Exercise Testing, American College of Cardiology/American Heart
Association Task Force on Assessment of Cardiovascular Procedures, Guidelinesfor Exercise
Testing, 8 J. AM. C. CARDIOLOGY 725 (1986).
95 Subcommittee on Coronary Artery Bypass Graft Surgery, American College of
Cardiology/American Heart Association Task Force on Assessment of Diagnostic and
Therapeutic Cardiovascular Procedures, Guidelines and Indications for Coronary Artery
Bypass Graft Surgery, 17 J. AM. C. CARDIOLOGY 543 (1991).
96 Committee on Pacemaker Implantation, American College of Cardiology/American Heart
Association Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular
1994]
Influence of a Professional Organization
97
established standards for anesthetic monitoring during surgery.
These guidelines were developed by panels of experts, based on published data and their own clinical experience.98
In general, studies have shown that simply developing and disseminating practice guidelines is not sufficient to change physician
behavior, even when there is widespread knowledge among physicians about the guidelines. For example, despite the efforts of professional societies to reduce the rate of cesarean sections, the rate
has remained high in both the United States and Canada. After
Canada's Society of Obstetricians and Gynecologists issued its practice guidelines for cesarean sections, roughly 90% of obstetricians
surveyed reported that they knew about the guidelines, and more
than 80% reported that they agreed with the guidelines.99 Yet two
years after the release of the guidelines, there was only a small decrease in the cesarean section rate.100 Indeed, if that small a decline
were multiplied over time, it would take more than thirty years for
Canada's cesarean section rate to reach the medically desirable
level.10
Similar results were found in a study of practice guidelines issued
by the National Institutes of Health ("NIH"). Between 1977 and
1986, the NIH developed guidelines on sixty different practice questions. 10 2 In a study of the impact of four of those guidelines (two that
applied to treatment of breast cancer, one that applied to cesarean
sections, and one that applied to coronary artery bypass surgery),
were largely unsuccessful in
researchers found that the guidelines
3
10
changing physician behavior.
The failure of information alone to change physician behavior is
not surprising. Sociological studies on the diffusion of innovation in
medicine, in agriculture, and in other settings have come to the
same conclusion: knowledge about and availability of an innovation
Procedures, Guidelinesfor Implantation of CardiacPacemakersand Antiarrhythmia Devices,
18 J. AM. C. CARDIOLOGY 1 (1991).
97 Ellison C. Pierce, Jr., The Development of Anesthesia Guidelines and Standards, 16
QUALITY REV. BULL. 61 (1990).
98 John T. Kelly & James E. Swartwout, Development of PracticeParametersby Physician
Organizations,16 QUALITY REV. BULL. 54, 56 (1990).
99 Jonathan Lomas et al., Do PracticeGuidelines Guide Practice? The Effect of a Consensus
Statement on the Practiceof Physicians, 321 NEw ENG. J. MED. 1306, 1308 (1989).
100 Id. at 1310.
101 Id.
102 Jacqueline Kosecoff et al., Effects of the National Institutes of Health Consensus
Development Program on Physician Practice, 258 JAMA 2708 (1987).
103 Id. at 2712.
[Vol. 57
Albany Law Review
600
are almost never
adequate by themselves to cause adoption of the
4
innovation.
10
There are a number of reasons why physician behavior does not
change by the mere dissemination of practice guidelines. Some studies suggest that the judgment of physicians may be shaped more by
their own clinical experiences than by nationally developed guidelines. 1 5 Physicians may also resist practice guidelines as an unwarranted intrusion on their decision-making authority. 0 6 Indeed,
some theoretical inquiries suggest that personal independence is a
fundamental element of professionalism in the United States. Burton Bledstein has described a "culture of professionalism" that developed in the 19th century and that rested on the idea of "selfgoverning individual[s] exercising [their] trained judgment.'
0
7
By
permitting professionals to be truly autonomous individuals, society
could fully realize the benefits of the professionals' creative
energies.' 08
Other factors underlying the resistance to change include personal
interests and patient preferences. For example, in the case of
cesarean sections, physicians may continue performing the procedure unnecessarily because they believe doing so will reduce their
risk of malpractice liability-a jury might mistakenly attribute a
newborn's pre-labor injury to the use of vaginal delivery. Physicians
may also be responding to other financial and personal incentives to
perform cesarean sections, primarily that cesarean sections are
more remunerative and require less time than vaginal deliveries.
Finally, physicians may be acceding to their patients' requests for
cesarean sections (which may stem from the patient's wish to avoid a
painful and prolonged delivery). 0 9
The failure of practice guidelines to change behavior cannot be attributed simply to the inability of physicians to modify wellentrenched practices. There are a number of cases in which the
medical profession has rapidly adapted to medical innovations, even
without the issuance of practice guidelines. For example, within five
years of its introduction in the United States, laparoscopic cholecys104 JAMES S.
COLEMAN ET AL., MEDICAL INNOVATION: A DIFFUSION STUDY 55
(1966).
See Peter J. Greco & John M. Eisenberg, ChangingPhysicians'Practices,329 NEw ENG.
J. MED. 1271 (1993); Louise Pilote et al., Return to Work after Uncomplicated Myocardial
Infarction:A Trial of Practice Guidelines in the Community, 117 ANNALS INTERNAL MED. 383,
388-89 (1992).
105
106 See Greco & Eisenberg, supra note 105, at 1273.
107 BURTON J.
BLEDSTEIN, THE CULTURE OF PROFESSIONALISM: THE MIDDLE CLASS AND THE
DEVELOPMENT OF HIGHER EDUCATION IN AMERICA 87
108 Id. at 91-92.
109 See Lomas et al., supra note 99, at 1310.
(1976).
1994]
Influence of a Professional Organization
tectomy has replaced more traditional surgical methods in roughly
80% of operations to remove the gallbladder. 110 This rapid adoption
of a new procedure is particularly striking given the unavailability of
any rigorous studies comparing the two procedures,"1 and the fact
different from
that laparoscopic surgery involves techniques very
2
those used in more traditional forms of surgery."
In some cases, practice guidelines have been successful in changing physician behavior, and the successes can be attributed to a
number of different factors. The successful implementation of stanof
dards for anesthetic monitoring was a result of a combination 113
mandates from both hospitals and licensing boards for their use,
and reductions in malpractice premiums that were conditioned on
4
their use."
Reimbursement policies of health care insurers can also be important. Before the issuance of guidelines on cardiac pacemaker implantation, data suggested that at least 20% of pacemakers were not
warranted. 1 5 Following the issuance of the guidelines, there was a
28% decline in the use of pacemakers in Medicare patients." 6 The
decline probably reflected the use of the pacemaker guidelines by the
Medicare system in deciding when to cover pacemaker implantations." 7 Tying reimbursement to adherence to practice guidelines is
an obvious method for achieving physician adoption of practice
guidelines. Physicians are not very likely to perform procedures for
which they are not compensated.
Indeed, the interesting question is why reimbursement has not
been predicated more frequently on physician adherence to practice
guidelines. For the most part, the answer probably lies in the fact
that most practice guidelines have been developed relatively recently and are available for only a small percentage of medical decisions. In addition, as indicated above, there is a good deal of
110 NIH Consensus Development Panel on Gallstones and Laparoscopic Cholecystectomy,
Gallstones and Laparoscopic Cholecystectomy, 269 JAMA 1018, 1018 (1993).
111 Id.
112 Barak Gaster, The Learning Curve, 270 JAMA 1280 (1993).
113 John H. Eichhorn, Prevention of IntraoperativeAnesthesiaAccidents and Related Severe
Injury through Safety Monitoring, 70 ANESTHESioLOGY 572 (1989) [hereinafter Eichhorn,
Prevention]; John H. Eichhorn et al., Standardsfor PatientMonitoring DuringAnesthesia at
HarvardMedical School, 256 JAMA 1017 (1986) [hereinafter Eichhorn et al., Standards].
114 Eichhorn, Prevention, supra note 113; Pierce, supra note 97, at 63.
115 Allan M. Greenspan et al., Incidence of Unwarranted Implantation of Permanent
CardiacPacemakers in a Large Medical Population, 318 NEw ENG. J. MED. 158, 160 (1988).
116 Janet B. Mitchell et al., The Medicare Physician Fee Freeze, HEALTH AFFAIRS, Spring
1989, at 21, 27 (1989).
117 Kelly & Swartwout, supra note 98, at 54.
Albany Law Review
[Vol. 57
resistance by the medical profession to the imposition of practice
guidelines. When Blue Cross and Blue Shield of Illinois disclosed its
plan to require physicians in its managed care networks to follow
practice guidelines, the AMA criticized the plan as an unwarranted
intrusion on professional judgment by an insurance company.'
Health care insurers may have resisted using their reimbursement
policies to impose practice guidelines in the belief that the benefits of
using practice guidelines did not outweigh the costs of antagonizing
physicians.
Strict regulatory oversight has also been cited as a mechanism for
ensuring adherence to practice guidelines. In a study of coronary
artery bypass surgery in New York State, researchers found a very
low rate of inappropriate operations. 1" 9 The authors of the study attributed the findings to the state government's careful regulation of
bypass surgery, including the requirement that hospitals satisfy
high standards of quality before they are certified or recertified as
centers for open heart surgery. 120 Similar results have been
achieved by federal regulatory oversight. After the federal government imposed strict guidelines for the use of antipsychotic drugs in
nursing homes, researchers found that there was a substantial de121
crease in antipsychotic drug use in Tennessee nursing homes.
In several cases, such as the use of antibiotics or cesarean sections, practice guidelines have been adopted by physicians when local "opinion leaders" (physicians whose opinions tend to be followed
by other physicians in their community) have adopted the guidelines
and encouraged their colleagues to do so as well. 1 22 The phenomenon of opinion leadership is widely recognized and studied in the
marketing and sociological literature. 23 Purveyors of consumer
products have long known that while media advertising is useful for
118 Michael L. Millenson, Blue Cross to Enforce Treatment Guidelines, CHI. TRIB., Nov. 10,
1993, at Al.
119 Lucian L. Leape et al., The Appropriateness of Use of Coronary Artery Bypass Graft
Surgery in New York State, 269 JAMA 753, 758 (1993).
120 Id. at 760.
121 Ronald I. Shorr et al., Changes in Antipsychotic Drug Use in Nursing Homes During
Implementation of the OBRA-87 Regulations, 271 JAMA 358 (1994).
122 Daniel E. Everitt et al., ChangingSurgicalAntimicrobialProphylaxisPracticesThrough
Education Targeted at Senior Department Leaders, 11 INFECTION CONTROL HoSp.
EPIDEMIOLOGY 578, 579 (1990); Jonathan Lomas et al., OpinionLeaders vs Audit and Feedback
to Implement Practice Guidelines: Delivery After Previous Cesarean Section, 265 JAMA 2202,
2206 (1991).
123 E.g., Kenny K Chan & Shekhar Misra, Characteristicsof the Opinion Leader: A New
Dimension, 19 J. ADVERTISING 53 (1990); Dorothy Leonard-Barton, Experts as Negative
Opinion Leaders in the Diffusion of a Technological Innovation, 11 J. CONSUMER RES. 914
(1985).
1994]
Influence of a Professional Organization
making people aware of a new item, consumers often turn to influential friends and acquaintances for guidance when deciding whether
to try the item. 1 2 4 The importance of opinion leadership for physicians was illustrated in a major study dealing with the adoption of a
new prescription drug in four Midwestern towns in the 1950s. Researchers found that the most important factor in explaining how
rapidly a physician adopted the drug was whether the physician was
well-integrated into professional and social networks with other
physicians. 1 25 Physicians who were professionally and socially isolated tended to be much slower to incorporate the drug into their
practice. 126
Opinion leadership reflects a number of factors. There will often
be a good deal of uncertainty about the appropriate use of a test or
treatment, and the greater the degree of their uncertainty, physicians, like other people, are more likely to turn to respected friends
or acquaintances for guidance. While the existence of nationally developed consensus guidelines suggests that the uncertainty has been
resolved and therefore there is little need to validate the guidelines
through local opinion leaders, it is also the case that consensus
panels are convened precisely because of a need to overcome major
differences of opinion among people in the field. The issuance of the
guidelines does not mean that the differences have been resolved. In
addition, there appears to be a common distrust of researchers by
practicing physicians. In one study, practicing physicians reported
that they viewed researchers as biased by their personal interest in
having their theories validated and their work published. 1 27 As a
result, the practitioners believed that scientific studies tend to exaggerate the value of a new therapy, and that the initial promise of an
innovation often does not hold up when the innovation is more
widely used. 12 Finally, on matters of ethics, where there are often
widely divergent views on what constitutes proper conduct, individual physicians are especially likely to trust their own judgment over
that of national experts.
In short, the mere issuance of guidelines by professional societies
rarely suffices to change physician behavior. Physicians often have
countervailing incentives to maintain their existing practices. Consequently, additional measures are needed, such as acceptance and
124 Chan & Misra, supra note 123, at 53.
125 COLEMAN ET AL., supra note 104, at 79-112.
126 Id.
127 Ann L. Greer, The State of the Art Versus the State of the Science: The Diffusion of New
Medical Technologies into Practice,4 INTL J. TECH. ASSESSMENT HEALTH CARE 5, 9-10 (1988).
128 See id. In many cases, of course, innovations do not live up to their initial promise.
604
Albany Law Review
[Vol. 57
encouragement by local opinion leaders, financial incentives, and/or
credible threats or methods of enforcement.
As indicated, for both the AMA's ethics guidelines and professional practice guidelines, the additional measures needed to ensure
adoption tend to come from outside, rather than from within the profession. Why has this been the case? First, there is a natural reluctance to engage in enforcement when the discipline is meted out to
colleagues. Members of a commission investigating police corruption in New York City came to a similar conclusion: police corruption
exists, despite such periodic investigatory commissions, because police are reluctant to clamp down when they discover corruption in
the ranks.1 2 9 Physicians, like other professionals engaged in selfregulation, can readily sympathize with ethical lapses of their peers
because it is easy to imagine themselves making similar errors.
Moreover, medicine has always been an unusually collegial profession. The Hippocratic oath 13 0 instructs physicians to give special
preference to the sons of their colleagues, and fealty to the profession
is considered as important in the oath as devotion to patients.
Self-enforcement is also weak because it is poorly funded. Physicians who serve on the disciplinary boards of their professional bodies do so without compensation. In addition, there is little money
available for staff, and the boards have no subpoena authority. Consequently, few cases can be pursued, and rigorous investigations are
not possible. Moreover, even when cases are prosecuted, there are
substantial financial risks to the professional society. Physicians
who are disciplined often challenge their sanction through timeconsuming and costly litigation. Indeed, the legal fees for defending
a case can deplete much of a small medical society's annual budget.
Antitrust liability is of particular concern with its potential for
treble damage and attorneys' fee awards.'
As the U.S. Congress
found when it enacted the Health Care Quality Improvement Act of
1986132 to provide physicians some protection against retaliatory
lawsuits, the threat of liability "unreasonably discourages physi33
cians from participating in effective professional peer review.'
129 Joseph P. Armao & Leslie U. Cornfeld, Why Good Cops Turn Rotten, N.Y. TIMES, Nov. 1,
1993, at A19.
130 See THOMAS A. MAPPES & JANE S. ZEMBATY, BIOMEDICAL ETHICS 53 (3d ed. 1991)
(reprinting the Hippocratic oath).
131 William J. Curran, Medical Peer Review of Physician Competence and Performance:
Legal Immunity and the Antitrust Laws, 316 NEW ENG. J. MED. 597 (1987).
132 42 U.S.C. §§ 11,101-11,152 (1988).
133 42 U.S.C. § 11,101(4).
1994]
Influence of a Professional Organization
CONCLUSION
As policymakers consider how to regulate the use of medical innovations by physicians, they should recognize the important differences between establishing and enforcing professional guidelines.
The medical profession's experience with ethics guidelines and practice guidelines indicates that society can rely on the profession to
develop responsible standards. In addition, principles of change
theory suggest that physicians will be more receptive to restrictions
on their autonomy if they are involved in the process of developing
the restrictions.'
However, on the issue of enforcement, reliable
mechanisms have come from outside the profession, generally in the
form of regulatory mandates or reimbursement policies.
An important caveat is in order. It is likely that the establishment
and enforcement of guidelines are related rather than independent
endeavors. Specifically, the willingness of the medical profession to
enact responsible guidelines might diminish if robust enforcement
mechanisms were in place. It may be that tough guidelines can be
adopted precisely because there often is little risk that they will be
enforced. On the other hand, the alternative of having outside
groups establish the guidelines might provide sufficient incentive for
the profession to continue developing rigorous ethical standards
even under a system of regular, reliable enforcement. How all this
would play out is indeterminate; whether enhanced enforcement
would diminish the zeal of professional standard-setting is an empirical question that can be resolved only by monitoring the regulatory
process and measuring the impact of greater enforcement activities.
134
Greco & Eisenberg, supra note 105, at 1272.